• No results found

Ductility in lightweight concrete with fiber

N/A
N/A
Protected

Academic year: 2022

Share "Ductility in lightweight concrete with fiber"

Copied!
102
0
0

Laster.... (Se fulltekst nå)

Fulltekst

(1)

Studiep Konstru

Forfatte Milad A Fagansv Veilede

Tittel på Duktilit Engelsk Ductilit Studiep Emneor Lightwe Ductilit Fiber re

DE

rogram/spe uksjoner og

er:

Ahmadyar varlig: Kjell er(e): Kjell T

å masteropp tet i Lettbeto k tittel:

ty in Lightw oeng: 30SP rd: Steel-fib eight concre ty,

einforced co

T TEKNIS

M

sialisering:

materialer

l Tore Fosså Tore Fosså

pgaven:

ong med Fib

weight Conc P

ber, ete, oncrete

SK-NATUR

MASTE

å

ber

crete with Fi

RVITENSK

EROPP

iber

KAPELIGE

PGAV

Vår

Mil

(

Sidetall: 8 + vedlegg

Stavanger

E FAKULT

E

r semestere

Åpen ilad Ahm

(signatur forfa

86 g

, 15.06.201 TET

t, 2011

madyar

atter)

1

(2)

Abstract

This master thesis presents the influence of different fiber in high-performance lightweight concrete and the ductility capacity of reinforced lightweight concrete beam. Twelve beams with length of 2.2m and reinforcement ratio 0.24 have been tested under 4 point bending, three of them were made by normal density aggregates as references beams. The target concrete compressive strength for all beams were 50MPa.

Three different types of fibers such as steel fiber, Polypropylene macrofiber and Polypropylene microfiber (PP-fiber)with two different fraction volume has been used to study the influence of fibers on high-performance concrete.

Two types end-hooked steel fibers with a length/diameter 50mm/1mm (N50) and 35mm/0.55mm (N35) has been evaluated. The behavior of plastic-fiber with a length/diameter 50mm/1mm and

Polypropylene fiber with length/diameter 12mm/0.22mm have been also investigated. Steel-fibers and plastic-fiber were studied by two volume fraction 1 and 2%. The volume fraction of PP-fibers were decided to be 0.4 and 0.6% because of low volume density.

Test results showed that use of plastic-fiber and PP-fiber has no influence on crack-width development.

Beams with plastic-fiber showed the same crack-width comparing to concrete beams without fibers.

PP-fiber give negative results specially by increasing the volume fraction from 0.4% to 0.6%. End- hooked steel fibers decreased the crack opening under loading compering to references beams. Test result showed that use of N50 in a volume fraction greater than 1% will decrease the crack-width at beam under same loading conditions comparing to beam with 1% volume fraction.

All beams failed due flexural failure under loading. The lightweight concrete showed brittle failure and sudden fracture. Beams with end-hooked steel fibers had a ductile behavior during failure, increasing in volume fraction of steel fiber from 1 to 2% increased the flexural moment capacity of beam under loading. Both type of steel-fiber with a 1 and 2% volume fraction showed the highest increasing in ductility and moment capacity during loading to failure.

Plastic-fiber resulted in to increasing ductility of beams, volume fraction greater than 1% had better influence on beam ductility and moment capacity of beam. The plastic-fiber beam with volume

fraction 2% failed at the same load as normal density concrete beam but with higher ductility capacity.

(3)

PP-fiber with higher volume fraction than 0.4% caused segregation by swallow cement pasta, and caused bleeding by releasing the water after casting of concrete. Bleeding and segregation caused low concrete compressive strength and low moment capacity for beam during loading.

(4)

Table of Contents

Abstract...1

I. Symbols...5

II. Preface...8

Chapter 1...9

1.1 Introduction to concrete...9

1.2 Types of concrete...9

1.3 Strength classes of concrete...10

1.4 Aggregate...10

1.5 Cement...11

1.6 Chemical Admixtures...13

1.7 Plasticizers/Superplasticizers...13

1.8 Pozzolan...14

1.8.1 Fly Ash...14

1.8.2 Silica Fume...14

1.9 Self-compacting concrete (SCC)...16

Chapter 2: Ductility of High Performance Lightweight Concrete ...18

2.1 introduction to Lightweight concrete ...18

2.2 High Performance lightweight aggregates Concrete...19

2.3 Lightweight self-compacting concrete...19

2.4 Mechanical properties of LWC...20

2.4.1 Compressive strength ...20

2.4.2 E-modulus...21

2.4.3 Cracking and shrinkage ...21

2.5 Non-steel reinforcement...24

2.6 Steel Fiber Concrete...25

2.6.1 Mechanical properties of steel-fiber concrete...27

2.7 Fiber bridging...28

2.8 Material qualities for steel fiber reinforced concrete (SFRC)...28

2.8.1 Calculation of the residual strength in fiber reinforced concrete(FRC)...29

2.9 Concrete Reinforced composites...31

2.10 Shear in concrete reinforced composites...32

2.11 Crack Width Analysis for normal reinforced concrete...33

2.12 Crack Width Analysis for fiber and conventional reinforcing ...34

2.13 Ductility of concrete ...36

2.14 Curvature ductility...38

2.15 Concrete fracture mechanical...41

2.16 Fracture energy in concrete...42

Chapter 3 : EXPERIMENTAL PROGRAMME...44

3.1. Mix Proportions...44

3.2 Fiber types...45

(5)

3.3 Specimen Details ...46

3.4 Mixing and casting...48

3.5 Beam testing...49

Chapter 4: Experimental Result ...49

4.1 Compressive strength...49

4.1.1 discussion ...51

4.2 Tensile strength ...51

4.2.1 discussion of split test result...55

4.3 E-modulus ...57

4.4 Stress-strain diagram ...59

4.5 Behavior of beam specimens...61

4.5.1 Lightweight concrete beam without fiber (LWC_UF)...63

4.5.2 Reference beam self-compacting concrete without fiber (SCC_UF)...65

4.5.3 Reference beams self-compacting and vibrator-compacting concrete (SCC_MF, VBB_MF) ...66

4.5.4 Beam specimens with steel-fiber N50 (LWC_N50) and N35 (LWC_N35)...70

4.5.5 Beam specimens with polypropylene microfiber M12(PP-fiber) and Plastic-fiber M50 (PF) ...73

4.6 Analysis of results...74

4.6.1 Proposal for crack width analysis ...76

4.6.2 Deflection of beam specimens...78

5. Conclusion ...80

6. References...82

(6)

I. Symbols

Latin upper case letters

Latin lower case letters

,min

Cross sectional area of concrete

Cross sectional area of tensile reinforcement Minimum cross sectional area of reinforcement Modulus of elasticity of normal weight concrete E Se

c s s c cm

A A A E

cant Modulus of elasticity of concrete E Young's modulus of the fibers

E Young's modulus of reinforced bar Length of fiber

Design value of the applied internal bending moment V D

f s f

Ed Ed

l M

esign value of the applied shear force V Fiber volume fraction

Crack width

f

wk

Effective depth of a cross-section Compressive strength of concrete

Design value of concrete compressive strength

Characteristisc compressive cylinder strength of concrete at

c cd ck

d f f

f 28 days

Mean value of concrete cylinder cylinder compressive stregth Characteristic axial tensile strength of concrete

Mean value of axial tensile strength of concrete Yield st

cm ctk ctm y

f f f f

,

rength of reinforcement

Design yield strength of reinforcement

The residual stress of fiber reinforced concrete Partial factor

Partial factor for concrete Partial fact

yd tk res

C S

f f γ γ

γ or for steel reinforcement Partial factor for fiber property Compressive strain in the concrete

Ultimate compressive strain in the concrete The main strain in the reinforceme

m c cu sm

γ ε ε

ε nt

(7)

Abbreviations

The main strain in the concrete First crack strain of plain concrete Strain of reinforcement

Compressive stress in the concrete Compressive stress in the concrete at

cm mu u

c

cu c

ε ε ε σ

σ ε

,max ,

Maximum stress of fiber

Avrage stress in all fibers bridging a crack

u fk

fk mid

σ σ

ACI American concrete institute EC2 Eurocode 2

FA Fly ash

FRC Fiber reinforced concrete LWC Lightweight concrete M12 Fiber with length 12 mm NDC Normal density concrete

Self

SCC -compacting concrete SFRC Steel fiber reinforced concrete VCC Vibrator-compacting concrete

N50 End-hooked steel fiber with ( / )50 /1 N35 End-hooked steel fiber with ( / )35 / 0.55 M50

f f

l d mm mm

l d mm mm

Polypropylene plastic fiber with ( / ) 50 /1 M12 Polypropylene microfiber with ( / ) 12 / 22

f f

l d mm mm

l d mm µm

(8)

Abbreviations of beams and concrete ID

Ref_SCC_MF Referance beam with self-compacting concrete with 2% N50 steel fiber Ref_VCC_MF Referance beam with vibrator-compacting concrete with 2% N50 steel fiber Ref_SCC_UF Referance beam with self-compacting concrete without fiber

Ref_LWC_UF Referance beam with lightweight concrete without fiber LWC_N50_2% Lightweight concrete beam with 2% N50 steel fiber LWC_N50_1% Lightweight concrete beam with 1% N50 steel fiber LWC_N35_2% Lightweight concrete beam with 2% N35 steel fiber LWC_N35_1% Lightweight concrete beam with 1% N50 steel fiber

LWC_PF_2% Lightweight concrete beam with 2% M50 polypropylene plastic fiber LWC_PF_1% Lightweight concrete beam with 1% M50 polypropylene plastic fiber LWC_PPF_2% Lightweight concrete beam with 2% M12 polypropylene microfiber LWC_PPF_1% Lightweight concrete beam with 1% M12 polypropylene microfiber

(9)

II. Preface

The Popularity of high rise buildings with great design has explored in the last two decades. The title of highest building does not last more than few years, since 1996 which Willis tower was highest building with 442m height, the title of highest building have been beaten 6 times. Today’s record holder is Burj Khalifa (828m tall). Use of normal concrete in such a high building will give a very high dead load which is a big challenge for design engineering. A secondary use of lightweight concrete which is 25 % lighter than normal concrete can reduce use of reinforcement bars and support work which leads to considerable cost saving and less dead load design.

Use of lightweight concrete specially high-performance lightweight concrete has been founded challenges in seismic zones because of low ductility capacity. The ductility has to be increased to satisfy the minimum seismic design requirements. Use of fiber-reinforced concrete can be one of solutions for improved ductility.

The purpose of this thesis is to study the influence of fibers on ductility behaviors in high-performance lightweight concrete. The funds for this thesis provided by Mechanical and Structural Engineering Institute of University of Stavanger. The author would like to extend his gratitude to his supervisor PhD. Kjell Tore Fosså for his time dedication and his support and guidance in the process of testing and writing this thesis.

The author would like to express his gratitude to Samdar Kakay, John Grønli and Einar Holven for laboratory and technical support during casting and testing of beams.

Stavnager, June 15th 2011,

(10)

Chapter 1

1.1 Introduction to concrete

Use of cement and concrete went back to 2000 year B.C. Roman Empire used quicklime, pozzolana and aggregate of pumice to build pantheon and Baths of Caracalla. The name “concrete” comes from Latin concretus, which means to grow together.

First modern concrete was made in 1756 by British engineer John Smeaton. He defined his concrete as mixture of hydraulic cement and pebbles as a coarse aggregate. The most important year for concrete industry is 1824 when Joseph Aspdin invented Portland Cement. He burned ground limestone and clay together. He knew that he can change the chemical properties of the ground limestone by burning it, the process create a stronger cement.

Today is concrete most used material in construction business. It is because of properties such as high compressive strength, durability, flexibility of shape and fire resistance. In 2009 produced worldwide cement factories about 2,800,000,000,000 tons Portland cement ref.[1].

Concretes strength mostly controls by w/c ratio, low w/c ratio give higher strength and lower permeability. Normally w/c ratio starts from 0.4 up to 0.8.

Hydration process starts between water and cement and result of this reaction is cement paste. Cement paste glues the aggregate together. Aggregate has normally higher strength compare cement paste but it depends on strength of concrete. In normal strength concrete will cement paste fail due compressive pressure. Hydration process is a critical phase for strength of concrete.

Concrete weakness is in tension, cement pasta will crack fast when concrete is under tension. To solve these problems used Joseph Monier in 1849 steel reinforcing bars. Today they use steel fibers, glass fiber, or plastic fiber to increase strength in tension.

1.2 Types of concrete

Concrete is divided into three categories based on unit weight. Normal weight concrete which contains natural sand and gravel or crushed rock aggregates, has a density around 2400 kg/m³. Lightweight concrete(LWC) which contains lightweight aggregates with lower bulk density, has a density around 1800 kg/m³. The third type of concrete is heavyweight concrete with a density higher than 2400 kg/m³.

(11)

1.3 Strength classes of concrete

Strength of concrete is classified into three strength classes. Strength lower than 20MPa knows as low- strength concrete. It is used in structures member such as floor-system, facade etc. Main reason of using low strength concrete is saving in project cost.

Moderate-strength concrete has a strength between 20 to 40MPa. It is also referred to as normal concrete and most used in structural work ref.[2].

Concrete with a strength higher than 40MPa is called high strength concrete or high-performance concrete. It is used for structures, offshore, oil and gas industry. In case of saving in material cost, it is possible to replace normal concrete by high-performance concrete to reduce the size of structural member and the use of reinforcement.

Table 1-1: Strength classes for concrete [Table 3.1 En1992-1-1:2004E]

1.4 Aggregate

Aggregate occupies between 65 and 75% of the concrete volume. Manufacturers produce usually aggregates with grading 0-8, 8-16 or 8-22. Optimized packing and minimal voids in concrete depends on how well aggregates are graded.

There are two types aggregates, coarse aggregate and fine aggregate.

(12)

The size of coarse aggregate is larger than 5mm. Fine aggregate is smaller than 5mm and made of sand and it helps to get better well graded aggregates and reduce use of the cement paste. Adding adequate fine material in a concrete mixture will avoid segregation and bleeding. Addition more than necessary fine material will make concrete sticky and unworkable.

Bulk density of aggregates determine the weight and category of concrete whether it is lightweight, normal or heavyweight concrete.

1.5 Cement

For describe cement in general sense, can say that it is a binder which used for adhesive other material together. Revolution in cement industry start in early years of industry age.

Cement divides in two categories, Hydraulic cement and non-hydraulic cement.

Hydraulic cement is also called water resistance cement or Portland cement. Non-hydraulic cements are gypsum and lime cement.

Portland cement is the most used cement in structural concrete. The main Portland clinker is limestone.

The limestone is mixed with bauxite, quartz and gypsum to provide the correct composition of oxides.

The production of Portland cement is depend on the resource and local availabilities to materials.

Limestone is a calcium silicates product and can be replace by chalk, marl and sea shell. They are all common material sources of calcium silicate. The Portland cement clinker also contains iron,

aluminates, and aluminoferrites of calcium.

Table 1-2: Oxides and clinker compound of Portland cement ref[2]

Abbrevation Oxid Name Compound Abbrevaition

C CaO 3 CaO • SiO2 C3S

S SiO2 2 CaO • SiO2 C2S

A Al2O3 3 CaO • Al2O3 C3A

F Fe2O3 Iron oxide C4AF

M MgO Magnesium oxide C4A3S

Calcium oxide, or lime Silicon dioxide, or silica Aluminium oxide, or alumina

4 CaO • Al2O3 • Fe2O3 4 CaO • 3Al2O3 • SO3

(13)

Table 1-3: Notation and mechanical properties of different cement ref[3]

Product Name Norcem Standard Norcem Standard FA Norcem Industri Norcem Anlegg Norcem Anleggsement FA

CEM I 42,5 R CEM II/A-V 42,5 R CEM I 42,5 R CEM I 52,5 N CEM II/A-V 42,5 N

Compressive strength Mpa

Setting time(min) 130(B)/125(K) 130 100 120 165

1day 21 21 32 18 13

2days 32 31 40 30 22

7days 42 40 49 46 35

28 days 52 52 57 60 53

Nation according NS-EN197-1

         

Figure 1-1: overview over the cement production process ref[4].

(14)

1.6 Chemical Admixtures

Admixtures are used as ingredients of concrete and are added to the batch immediately before or during the mix operation. The purpose of using admixtures is for change the properties of concrete such as increasing the strength, workability, plasticity, durability of concrete. Admixture can be used for reducing water demand and accelerating the cement hydration reaction.

Types of admixtures can be summarized such as:

• Accelerating admixtures

• Air-entraining admixtures

• Water-reducing admixtures

• Superplasticizers

• Polymers and latexes

• Antifreeze admixtures

• Anti-washout admixtures for underwater concreteing

• Anti-shrinkage admixtures 1.7 Plasticizers/Superplasticizers

Superplasticizers admixture is most used admixture in Norway, approximately 95% of all admixtures sold in Norway[7]. This admixture has ability to reduce three to four times the mixing water in given concrete mixture compared to normal water-reducing admixtures[2]. Workability and slump retention would be approved by adding superplasticizers admixture at batching plants without any refreshing of the concrete at work site. Adding of superplasticizers requires experience, by overdosing will concrete start segregation and can be described as like as soup.

Figure 1-2: overdosing of superplasticizers admixture

(15)

1.8 Pozzolan

Pozzolan is type of active mineral additions which can be added to cement in manufacturing or during concrete mixing. The important effect of pozzolan is increasing the long-term strength of concrete.

There are two types of pozzolans used in Norway, Fly Ash and Silica Fume.

1.8.1 Fly Ash

Fly Ash is added directly under cement production, Norcem produces two types FA cement in Norway, NORCEM STD-FA with 20% FA and ANLEGG FA with 30%FA.

Particle size of FA is around 10-20μm which is the same size as Portland cement.

FA in Portland cement will increase the long term strength of concrete and reduce cement consumption.

Anlegg-FA cement is usually used for producing high strength concrete. The benefits of Anlegg-FA can be named such as:

• Reducing alkali contents in concrete

• Low temperature development

• Excellent workability and permeability

• reducing risk of bleeding

Compressive strength of concrete with FA will be low at early age (3-7days) because of low temperature development.

FA is an industrial byproduct from Coal industry and since there is no 1.8.2 Silica Fume

Silica Fume has a particle size of 0,1μm and it is a byproduct of smelting process of silicon metal (SiO) and ferrosilicon alloys. SF has become a very efficient filler because of the small size and it effect directly on fresh concrete properties. The benefits of using SF in concrete is depend on volume percentage of cement replacing which is normally up to 20%. ref[5]

(16)

Benefits can be summarized such as:

• increasing compressive Strength

• increasing workability

• reducing permeability

• reducing reinforcement corrosion

• more coherent

• eliminate bleeding

• homogenous concrete

• increasing frost resistance

• finer pore structures

Figure 1-3: Effects of increasing silica fume.ref[6]

(17)

1.9 Self-compacting concrete (SCC)

Durable concrete structure requires professional compaction by skilled workers. The reduction in the number of skilled workers and increasing the cost of construction's site in Japan, made Japanese engineers to find a new concreting method in early 1980s. The research led them to self-compacting concrete (SCC) which is a durable concrete and can reach every corner of a form-work without vibrating compacting. SCC is made of exactly same material as normal concrete with different in matrix volume, often higher than 340 l/m³. SCC is a matrix dominated concrete.

SCC must have high deformability and good resistance to segregation between coarse aggregate and mortar when the concrete flows through the confined zone of reinforcing.

NS-EN 206-9:2010 is the European standard for SCC and requirement for slump-flow has been given in 3 classes which shows in table 1-4.

Table 1-4: Slump-flow classes [table 1:NS-EN 206-9:2010]

Water/binder (w/b) ratio in SCC is normally lower than 0.45 which gives a matrix dominated concrete.

With higher w/b ratio than 0.45 is necessary to change the concrete composition for increasing the matrix volume. Increasing in matrix volume can cause separation because of low viscosity in matrix, it need more attention to keep matrix viscosity high enough to avoid separation.

Norwegian experience recommended to keep w/b ratio lower than 0.45 and higher than 0.3. Low w/b ratio can cause poor stability, more plastic shrinkage and drying shrinkage and gives higher strength than was planned. With w/ b ratio lower than 0.3 will concrete become very tough and flows heavily because of high matrix volume. [7]

o

Class Slump-flow in mm

SF1 550 to 650

SF2 660 to 750

SF3 760 to 850

(18)

to aggregate ratio. Adding silica fume in SCC gives a better stability but it is depend on the amount of addition.

Benefits of using SCC can be summarized such as:

• Decrease construction's time

• Improved durability

• Better surface finish

• Reduction of site manpower

• Better safety working environment

(19)

Chapter 2: Ductility of High Performance Lightweight Concrete

2.1 Introduction to lightweight concrete

The density of normal concrete is 2400 kg/m3, with replacing normal aggregate with lightweight aggregate will density of concrete decrease to 1750 kg/m3. Reducing of dead load up to 35%, will allowing greater design flexibility, reducing size of structure, longer spans and decrease reinforcing steel bars. The density class of LWC is showed in table 2-5.

Table 2-5: Density class for LWC

There are two types lightweight aggregate, manufacture which are produced by expanding products such as Vermiculite or perlite, calcining, blast-fumace slag, clay, fly ash, shale or slate. Second type is from natural materials, such as pumice, tuff or scoria.

The shape of particles can be rounded, angular or irregular. It is depend on source and production process. The lightweight aggregate which are produced in Norway have rounded shape and called Leca. Lightweight aggregates is barley used in structural concrete in Norway. The production of Leca is depending of each projects.

The southwestern Bell Telephone office building [8]in Kansas City is a great example of how effective lightweight concrete can be. Building was built originally 14 storeys structure but in 1928 the owner wished to add more storeys on the same foundation. After study the building they found out that foundation would support eight more storey with using normal density concrete. But if they replace concrete with lightweight concrete would be able to add 14 storeys on the same foundation. That was start of using lightweight concrete. The first structural lightweight concrete was built in 1929 in St.

Louis, they designed both frame and floor systems by lightweight concrete.

Density class 1 1.2 1.4 1.6 1.8 2

801-1000 1001-1200 1201-1400 1401-1600 1601-1800 1801-2000 Density(kg/m³)

(20)

2.2 High performance lightweight aggregates concrete

The proportioning of high performance lightweight concrete is based on cementitious material content such as cement type, chemical admixture and pozzolanic. The recommended slump is between 76 to100 mm for workability to be proved ref.[18].

The water/cementitious ratio is essential for the compressive strength and need extra attention. Because of larger cellular structure of lightweight coarse aggregate, it is important to remember that it has a high degree of water adsorption (up to 20%, depend on aggregates type). In case of Leca it is 10%

water absorption in first 24 hours and up to 25% after 28 days ref[appendix1a].

2.3 Lightweight self-compacting concrete

The trial batch tests by Ben C. Gerwick in California [9] showed that lightweight concrete with a slump flow below 430 mm is not likely to have full compacting performance. Lightweight concrete with a slump flow over 660 mm often leads to severe segregation. Slump flow in the range of 460 mm to 560 mm provided the required workability[9]. European standard for SCC [NS-EN 206-9:2010] does not applies to lightweight SCC because of limited experience. However, it requires a slump flow between 550 to 650 mm for class SF1 as it showed in table 2-5 ref [NS-EN 206-9:2010].

The most challenge to produce self-compacting lightweight concrete is the workability, low density of lightweight aggregates make flowability and cohesion low. For concrete become self compacting it needs to be able to flow under the own weight without any need for vibration or additional compaction.

Separation of concrete depends on stability of concrete. Stability increases with high flow resistance of the matrix. Higher workability decrease flow resistance, which results decreasing in the stability and risk for separation.

Flow resistance of the matrix λQ determined by FlowCyl-tests, experience shows when flow resistance become larger than approximately 0.75 may concrete become too viscous or tough which will cause low flowability. Flow resistance lower than approximately 0.5 may cause concrete bleeding and segregating. Ref.[7]

The shape and size of lightweight aggregate affects the workability and strength of concrete. Leca which is manufactured in Norway, have rounded particle shape and the density is lower than angular aggregate which results lower strength but better workability.

(21)

The total cementitious materials content should be approximately 550 kg/m³ based on researches has been done on self-compacting lightweight concrete. ref [9]

2.4 Mechanical properties of LWC

2.4.1 Compressive strength

Normal aggregates has a higher strength and stiffness than cement pasta in normal concrete, which means that aggregate carry most part of tension in concrete. That will cause a tensile stress between mortar and aggregate which will occur micro-crack after a while. When aggregates capacity exceeded will tension in micro-cracks become larger and cracks grow. In that phase concrete capacity is

exceeded and crack go through cement pasta. ref[10]

However, lightweight aggregate and mortar has almost the same stiffness, and produce a higher elastic compatibility leading to reduces stress concentrations in the mortar-aggregate interface. Low w/c ratio and using of admixtures will make sure a good mortar-aggregate interface which can also prevent bleeding. When tensile stress in LWC become larger, lightweight aggregate can not handle the stress and it will cause a suddenly crack and fracture which will go through mortar and aggregates. That make compressive strength of LWC a function of lightweight aggregate and cement paste strength and

stiffness.

When strength of concrete become larger than lightweight aggregates strength, mortar has to carry most part of concrete tensile stress. For high-performance LWC increases the mortar stiffness and strength.

Table 2-6: Strength classes for LWC table 11.3.1 EC2

12 16 20 25 30 35 40 45 50 55 60 70 80 13 18 22 28 33 38 44 50 55 60 66 77 88 17 22 28 33 38 43 48 53 58 63 68 78 88 flck(MPa)

flck,cube(MPa)

flcm

(22)

2.4.2 E-modulus

LWC has a linear loading curve and faster fall curve compare to NDC. LWC give more strain in maximum stress than NWC which cause lower E-modulus compare to NDC. Low E-modulus and higher strain in LWC will prevent shrinkage. ref[10]

2.4.3 Cracking and shrinkage

The most critical time for concrete is when concrete starts hardening. Most properties of concrete develops during this phase. Hardening process is between 1 to 2 weeks and it start between 6 to 12 hours after casting.

C-S-H adsorbs amount of water, when a hydrated cement paste is under sustained stress will C-S-H loss a most part of physical adsorbed water and it cause a creep strain in cement paste.

Reducing of cement paste volume during hardening of concrete will cause cracking and shrinkage in concrete ref.[5].

High strength concrete has a tendency to develop early age cracking. Early ages defines as minutes after casting when the concrete is in plastic age and some hours after casting when concrete is in early hardening age. The typical cracking problem will be caused of plastic settlement and plastic shrinkage.

Figure 2-4: Stress-strain diagram LWC & NDC

(23)

It has been assumed that LWC crack development behaves the same as NWC in early ages. It should be noted that Leca which has a particle density of 1300 kg/m³ have a water absorption about 40 liter per m³ concrete, which corresponds to a volumetric shrinkage of 40 000 μstrain. ref[11]

The creep coefficient φ for LWC is equal to creep coefficient φ of normal density concrete (NDC) multiplied by (ρ/2200)².

The creep strains is the same as NDC by strength classes higher than LC16/18, but the final drying shrinkage value for LWC increases 20% compare to NDC.[ref.EC2 11.3.3]

Type of shrinkage can summarized as:

• Plastic shrinkage: It occurs in fresh phase of concrete. The cause is water evaporating from the surface of the concrete. Plastic shrinkage can be expected in hot summer days and by strong wind.

• Plastic settlement: when concrete start bleeding in plastic phase, the solid particles move downward and water stays on top and cover the surface. The mechanism is as same as plastic shrinkage but in higher degree and depth.

3

3

,0

is a coefficient for determining creep coefficient 1.5 for 16 /18

1.2 for 16 /18 In case of relative humidity 40%:

The drying shrinkage values for NDC 0.46 The drying shri

lck lck

ooo cd

f LC

f LC

η η

ε

 ≤ 

=  ≥ 

=

nkage values for LWC εcd,0= 0.46 1.2 0.552× = ooo

(24)

• Autogenous shrinkage: Hydration process starts when cement react with water. The absolute volume will decrease when cement paste is in a liquid state. The macroscopic volume change occurring with no moisture transferred to the exterior surrounding environment cause

Autogenouse shrinkage which is an external volume change ref[24].

• Chemical shrinkage: It starts at same time with autogenous shrinkage. Pores in the binder phase start growing because of volume loss, which also increases air volume in the capillary pores.

That means that shrinkage turns into air voids within hardened cement paste. Chemical shrinkage is an interval volume reduction ref[24].

• Drying shrinkage: The mechanism is the same as plastic shrinkage but it expected in a stiff concrete. The shrinkage cracks on surface start after evaporation of surface's water and the pressure of negative pore develops crack and shrinkage on surface of concrete.

Figure 2-8: Drying shrinkage ref[15]

Illustration 2-7: Reactions causing autogenous and chemical shrinkage. [24] C = unhydrated cement, W = unhydrated water, Hy = hydration products,and V = voids generated by hydration.

(25)

2.5 Non-steel reinforcement

Non-steel reinforcement can be categories as fiber reinforced polymer(FRP) which can also called fiber reinforced plastic(FRP) and Glass reinforced plastic(GRP).

Plastic fiber were used since 1960s in some structure elements but not as main reinforcement bar. The important development of plastic fiber started in 1980s. The new development of medical equipment such as magnetic resonance imaging (MRI) become main reason for focus on use of plastic fiber as main reinforcement bar in concrete. MRI can not tolerate the presence of any steel reinforcement.

Engineers found a solution which were using of glass fiber reinforcement for surrounding and supporting MRI.

Environmental and chemical attacks become new consideration for using glass fiber as composite reinforcement in construction such as sea walls, industrial roof decks, chemical industry floor slaps and reactor equipment.

The major consideration in using plastic reinforcement as main reinforcement is the bond between fiber rods and concrete. The transformation of stress from the reinforcement to concrete is depend mainly on the bond interaction between those two materials.

Four types of bond failure are identified ref[18]:

1. Shearing off failure of the road windings

2. Local mechanical frictional pull-out( the rods are ribbed, and frictional resistance covers only a part of the surface)

3. full frictional pull-out where failure resistance is long the entire rod length 4. Plain frictional resistance in isolated strand rods

These four types of bond failure shows that the bond strength is depend on rod surface configuration and deformation.

(26)

Benefits of using PP fiber:

• Increasing fire resistance

• Increasing concrete density and durability

• Reducing amount of cracks from plastic shrinkage forces

• Reducing risk of separation/bleeding

• Faster and better construction Type of non-steel reinforcement:

• Polyolefin

• Polypropylene (twisted, fibrilated and monofilament)

• PVA

• Nylon

• Ar-glass

• Wollastonite 2.6 Steel fiber concrete

Concrete has high compressive strength but very low tensile strength capacity. Steel reinforcing bar carry the tensile strength in concrete and it prevent development of cracks under loading. Concrete structural members under loading develop crack before reinforcement take over the tensile forces. The cracks occur because of low tensile strain capacity of concrete. Cracks will reduce service life of concrete by reducing durability of concrete. Low durability will reduce the ability of concrete to resist weathering action and chemical attack. Development of cracks could be limited by using of steel-fiber.

Steel-fiber has high tensile strength and would take over the concrete tensile stress at early crack development phase. Fibers over the crack in concrete will decrease moment, shear and punching resistance. When a crack start growing, the fibers will transmit tensile force over the crack into the surrounding concrete ref.[17].

Steel bar reinforcement can be replaced by steel fiber and be added direct into the concrete at batching plant. Steel fiber-reinforced concrete(SFRC) is most used for slabs on grade, it reduce work time on construction and it reduce cracks compare to steel bar reinforcement.

(27)

The quality of SFRC is depend on how good fibers are distribute in concrete.

Self compacting concrete improve the homogenous distribution of fiber in concrete. Vibrator

compacted concrete (VCC) may cause poor distribution of fiber and give critical and weak points in concrete.

The volume fraction of fibers added to a concrete mix is measured as a percentage of the total volume of the concrete and fibers. The respect ratio of fiber is measured by dividing fiber length by fiber diameter.

Volume percentage of steel fiber is usual between 0.2%(20kg/m³) to 2%(157kg/m³). By adding more than 2% volume fraction, may cause poor concrete workability and reducing fiber dispersion.

Figure 2-9: Type of steel fiber ref[17]

(28)

2.6.1 Mechanical properties of steel-fiber concrete

The steel-fibers have little or no effect on compressive strength and E-modulus of concrete, in some cases it may increase the compressive strength but it is not mentionable. The E-modulus and

compressive strength should be determined experimentally when volume fraction of steel-fiber is greater than 1%.

Tensile strength of concrete increase by adding steel-fiber ref[16]. The effect of fiber on direct tensile strength is depend on the direction and distribution of fibers. Measurement of the direct tensile strength of SFRC is not given in any standard and it has no agreement how it can be tested by experiment.

Earlier researches where distribution of fiber were randomly, showed that the tensile strength increase up to 60%, but it is depend on how homogenous is the distribution ref[16].

Steel-fibers have most effect on the flexural strength of concrete, increasing up to 100% has been reported by earlier experiments ref[16]. Aspect ratio of the fibers and the fraction volume of fibers are important parameters for increasing the flexural strength.

The post-crack tensile strength of SFRC is depend on the distribution and orientation of fibers, as earlier been mentioned the casting process, workability and the size of specimen are important parameters.

Figure 2-10: The influence of steel fiber content on tensile strength

(29)

2.7 Fiber bridging

The main reason of using steel fiber is the increasing the toughness and energy absorption capacity of concrete which can lead to a greater ductility. Fibers bridge across crack and prevent it from growing and slow down the development of cracks opening in the concrete. Fracture or pull out of the fibers are depend on bond strength between concrete and fibers. For concrete become more ductile is best that fibers pull out, that means enough bond strength is important for ductility in concrete. The bond between fibers and concrete is primary decided from friction between the fibers and concrete. The shape of fiber is important parameter for the bond strength and amount of energy observation. For example end-hooked fibers give better friction and bond strength because of deformed ends. The end hooked needs to bend and yield before it can be pulled out of concrete. For pulling out a hooked steel fiber requires a work that is four times greater than for a smooth steel fiber ref[21].

For avoid fiber rupture the yield capacity of the fiber has to be sufficient. Pull out failure need lot of energy to take place which result a higher ductility ref[21].

Neutral axis in concrete beam section will move toward the compressive face when crack in concrete start growing. This movement of neutral axis become less and equilibrium by fibers pullout resistance and fibers bridging the cracks ref[22].

2.8 Material qualities for steel fiber reinforced concrete (SFRC) Norwegian preliminary guidelines (NPG) describe the residual stress as:

“The residual stress of SFRC based on assumption that steel fibers crossing a concrete crack

contributes to the tension capacity of the reinforces concrete in the same way as for reinforcement bars.

There will be no main change of direction of the fibers direction at the crack, the another assumption is that maximum force in a steel fiber at a crack is defined by the fiber's anchorage capacity” The residual stress of SRFC is showed in figure 2-11and it is given by NPG:

(30)

where

vf Fiber volume ratio (fiber volume)/(concrete volume)

σ fk ,max Maximum stress of fiber with anchorage length lb = lf /2 at a crack, decided by bond and upper yield limit

σ fk ,mid Average stress in all fibers bridging a crack, with random embedded length and orientation,

based on experience and experimental, SINTF recommended 500MPa for steel fiber and 250MPa for PP fiber

η1 Aspect ratio,σ fk ,mid fk ,max

η0 approximate 1/3 for randomly 3D distribution and orientation of fibers,1/2for fibers in planes parallel to tension direction

2.8.1 Calculation of the residual strength in fiber reinforced concrete(FRC)

Residual strength of steel-fiber and PP-fiber in case of FRC has been calculated based on NPG formulation, this calculation is for steel-fiber with 2 % volume fraction ref[17]:

, 0 ,

, , ,

,

/ 0.02

500

This value is given by SINTEF, it is based on experimental result.

1/ 3 0.02 500 3.33 / 3.33 /1.55 2.15

tk res f fk mid

f f c

fk mid fk mid tk res

ftd res tk m

f v

v V V

f MPa

f f MPa

η σ

σ σ

γ

=

= =

=

= × × =

= = =

Figure 2-11: Typical tension behavior for SFRC ref. [17]

(31)

The residual strength for steel-fiber will be 2.15MPa. The residual strength in PP-fiber decrease approximate half part residual strength than steel-fiber, calculation shows residual strength for plastic fiber with 2% volume fraction:

Strain and Stress distribution over a SFRC cross-section has been described in (NPG) as showed in figure2-11.

The stress and strain illustration is only for concrete with fiber reinforcement FRC. It has to be noted that when residual stress ftk,res become larger than 2.5 N/mm², the cross-section's compression zone height has to be determined.

, 0 ,

, ,

,

/ 0.02

250

1/ 3 0.02 250 1.67 / 3.33 /1.55 1.1

tk res f fk mid

f f c

fk mid tk res

ftd res tk m

f v

v V V

MPa

f MPa

f f MPa

η σ

σ

γ

=

= =

=

= × × =

= = =

Figure 2-12: Strain and Stress distribution over a SFRC cross-section ref[17]

(32)

Here p presents scale factor for load carrying capacity of SFRC which is depend on dimension of beams and e is yield factor for residual strength for self-compacting concrete (SCC).

p=1,1-0,7h>0,75

e=0,9 in upper parts of the element for SCC e=1,2 in lower parts of the element for SCC

e=1 when the residual strength is determined by bending tests of beams made of SCC 2.9 Concrete Reinforced composites

Definition of a composite material has been defined in ASTM 3878-95c as:

“Composite Material. A substance consisting of two or more materials, insoluble in on other, which are combined to form a useful engineering material possessing certain properties not possessed by the constituents.”

The difference between fiber reinforced concrete (FRC) and concrete reinforced composite is that FRC has no steel bar reinforcement in concrete.

NPG requires that all structures with safety level 2 or higher has to used reinforced composites, to make sure that conventional bars transfer all external forces in addition to the fibers. Because of using conventional bars all safety factor for fiber can be changed to γm=1.

The requirement is because of limited experience with fiber reinforced concrete.

Strain and stress distribution over a cross-section reinforced with both reinforced bar and steel fibers is showed in figure 2-13.

Figure 2-13: Strain and stress of concrete reinforced composite ref.(17)

(33)

By removing the safety factor, will the residual strength of concrete reinforced composite for steel fiber with 2% volume fraction be 3.33Mpa and for plastic fiber with 2% volume fraction be 1.67MPa.

2.10 Shear in concrete reinforced composites

Shear strength increase with using steel fibers in concrete. Calculation of shear strength in SFRC can be determined by:

Shear strength because of conventional reinforcement can be determined by EURO CODE2:

Total concrete's shear strength :

Shear capacity of beam with SFRC by volume fraction 2% will increase 34% compare to a beam without fiber. [apandix2]

Diagonal cracks can be reduced and controlled by adding stirrup reinforcement in beam. Reason of using stirrup reinforcement is to carry the redistributed shear stresses, mainly through tension after the formation of diagonal cracks. Tension will be transferred back to the concrete. [21] This will cause

0,8 ,

fd ftd res

V = f × × ×p b d

sw ,

A cot

Rd s s ywd

V = × ×z f × θ

d cd Rd s, fd

V = V + V + V

, ,

ftd res tk res

f = f

Figure 2-14: Shear forces on beam ref[17]

(34)

2.11 Crack Width Analysis for normal reinforced concrete

Theoretical analysis of crack width of beam with a bar reinforcement is given in EC2 7.3.4 as:

Sr,maxpresent the average final crack spacing.

where

r,max 3 1 2 4 ,

s = k c k k k× + ×φ ρ/ p eff

,max( )

k r sm cm

W = s ε − ε

1

2

3

is bar diameter

c is the cover to the longitudinal reinforcement 0,8 for high bond bars

1,6 for bars with an effectively plain surface 0,5 for bending

1,0 for pure tension 3, 4

k k k φ

 

=  

 

 

=  

 

= ; k4 = 0, 425

,max sm cm

,

, ,

maximum crack spacing

the mean strain in the reinforcement the mean strain in the concrete

(1 )

0,6 the stress in the tens

r

ct eff

s t e p eff

p eff s

sm cm

s s

s

s

k f

E E

ε ε

σ α ρ

ρ σ

ε ε

σ

− +

− = ≥

e

2

s 1

,

c,eff

, , ,

1

ion reinforcement assuming a cracked section is the ratio /

A '

A

; min{2,5( ); ( ) / 3; / 2}

0,6 for short term loading 0, 4 for long term loadi

cm p p eff

c eff c ef c ef

E E A

A b h h h d h d h

k α ρ ξ

α

= +

= × = − −

= ng

 

 

 

(35)

Table 2-7 shows the maximal crack width in all exposure class.

Table 2-7: Values of wmax(mm)(table 7.1N EC2)

It is important to be noted that crack spacing in SFRC is depend on bond between concrete and steel fibers. If steel fiber's bond with the concrete is stronger than permitted will it cause brittle crack on steel fibers and if bond is poor, fiber will have no effect.

2.12 Crack Zidth Dnalysis for fiber and conventional reinforcing

Analysis of mechanical behavior of concrete become complicated by adding fiber to concrete mix.

There are parameters such as fiber substrate, fiber geometry, fiber volume concentration, type of fiber, the distribution and orientation of fibers which are important to be in consideration in analysis of concrete members. The problem is that before adding fibers, a concrete structure member has already lot of design parameters such as stiffness, concrete strength, conventional reinforcement type, structural geometry and reinforcement type which it can be difficult to connect all those design parameters.

The orientation and distribution of fibers are depend on workability and type of concrete such as self compacting concrete or vibrate compacting concrete. Use of vibrator can have negative effect on

(36)

need to drive a micromechanical to find the stress-crack width relationship. [23]

There are few guidelines for crack width analysis for concrete reinforced composite, RILEM TC 1622- TDF is one of few guidelines for determine the crack width. The problem with this formulation is that it does not take the effect of fibers volume fraction and rotation in consideration.

As it showed in formula Sr,max multiplied by new parameter which take fiber length and diameter in perspective not volume fraction. The new parameter has to be equal or less than 1, this is depend on the aspect ratio of fiber. This parameter shows that use of fiber will decrease the crack spacing which will cause increasing the number of cracks and decreasing of crack opening.

To be noted that steel fiber with aspect ratio 50 will have no effect on crack spacing based on this parameter. Polypropylene microfiber (PP-fiber) have a high aspect ratio which cause error in this calculation.

,max 1 2

1 2

50 0, 25 50

/

k is coefficient which takes account of the bond proerties of the bars k is coefficient which takes account of the form of the strain distribution

is

r

eff f f

S k k

L φ

ρ φ

φ

   

=  + × ÷ ÷  × ÷÷

s c.eff s

c.eff

bar size

is effective reinforcement ratio A A

A is the area of the reinforcement contained A is the effective tension area

L is the fiber length is the fiber diameter

eff

f f

ρ

φ

(37)

COIN-rapport (may 2011) has present new guidelines for calculation of crack width for fiber-

reinforced concrete. Formulation is principle identical to EC2 7.3.4 (section 2.11) except addition of a factor k5 to calculation of maximum crack spacing. This factor takes the fiber additive into the

consideration.

Calculation of the residual strength in fiber reinforced concrete is showed in section 2.8.1.

2.13 Ductility of concrete

Ductility is an important property in reinforced structure in seismic zones.

It allowed the structure to deform when the maximum bearing capacity is exceeded and response to in- elasticity in server earthquake. Ductility may be defined as the ability to undergo deformation without a substantial reduction in the flexural capacity of the member (Park & Ruitong 1988). High ductility will cause high deflection when structural member were loaded until failure which will be a warning before total collapse.

,max

,max 3 1 2 4 5

,

5 , 2,5

( )

(1 / )

k r sm cm

r

s eff

ftk res ctm

w s

s k c k k k k

k f f

ε ε

φ ρ

= −

= × + × × × ×

= −

(38)

The ductile failure is initiated by yielding of steel reinforcement, concrete will crush when steel reinforcement begins to plastic deformation and loss the carrying properties.

Ductility is measured in terms of strain, displacement and rotation. [ref.18].

Strain ductility:

Curvature ductility:

Displacement ductility

For inelastic behavior to be sustainable it should all these ductility factor be grater than 1,0.

Concrete is known as a brittle material with low tensile strain capacity and poor fracture toughness.

With increasing concrete strength expecting more brittleness and decreasing in the concrete strain, ɛcu. In this study will get a concrete strain:

is maximum sustainable strain is the yield strain ductility

e y

y

µ ε ε ε ε

=

is the maximum sustainable curvature is the yield curvature

m y m y

φ φ

µ φ

φ φ

=

is the sum of yield displacement and plastic displacement µ = ∆y

4

58

2.8 27[(98 ) /100] 0, 0035

cm

cu cm

f ε f

=

= + − =

(39)

In case of concrete reinforced with both steel fiber and reinforcement bars the maximum strain of steel can not exceed 2.5‰(0.0025) ref[17]. This limitation give a higher compression zone and lower internal torque arm. By increasing the maximum strain in reinforcement will capacity increase too.

Brittle behavior of high-performance concrete make lot of engineers skeptical to use it in seismic region. It should be mentioned that ductility is not only depend on strength of concrete. Most part of the ductility performance dependent on reinforcement details and other parameters such as the material characteristics of the concrete, the geometry of structure, the material characteristics of reinforcement and the amount of longitudinal compressive reinforcement.

High strength concrete improved the bond between the reinforcement and concrete which can limited the deformation capacity of reinforcement and decrease ductility.

Ductility can increase by adding compression reinforcement bar in beams. In the case of seismic zones, concretes member should have better ductility performance than normal ductility. It is required more analysis on each structural member geometry and reinforcement design specially when high strength concrete is used.

For seismic zones is important that failure in structural member occurs by ductile flexural failure, instead of shear brittle failure. For avoid shear brittle failure it need to increase shear resistance of structural members by adding secondary shear steel reinforcements bar.

Ductility has relationship with crack of concrete, when cracks occur in concrete because of overloading will fibers bridge the cracks, in that way fibers decrease crack width and increase ductility of concrete . 2.14 Curvature ductility

Curvature ductility (μΦ) describes the ductility of reinforced concrete section and it is a ratio between curvature of concrete strain when it reach the ultimate strain and curvature of yield strength of tension reinforcement at start point of yielding. The moment and curvature at first reinforcement yield can be determined by assuming an under-reinforced section which has been suggest by Park and Paulay

(40)

The curvature at ultimate of reinforced concrete section can be calculated by finding the maximum value of concrete strain at the compressive fiber ref[20].

The value of concrete strain for normal concrete and lightweight concrete has been given in EC2 which is 0.0035.

''

''

2 2

(1 ) (1 )

where

is the distance from centroid of compressive forces in the steel and concrete to the centroid of tensile

( ') 2( ' / ) ( ')

is the tensil

y s y

sy y

y

s

M A f d

f

k d E k d

d

k n d d n n

φ ε

ρ ρ ρ ρ ρ ρ

ρ

=

= =

− −

= + + + × − +

e reinforcement ratio / ' is the compression ratio ' /

s s

A bd A bd

ρ

'

1

'

1

0.85 ( ) ( ')

2

where 0.85

is the depth of the equivalent rectangular stress block

u c s y

c c

u

s y s y

c

M f ab d a A f d d

c a

A f A f

a f b

ε ε β

φ

β

= × − + −

= = ×

= −

×

(41)

When the steel tension increase, will both k and a increase which will result a higher Ҩy and a lower Ҩu . That means decreasing the ductility. According to curvature formulation will ductility increase by increasing the concrete strength because k and a will decrease.

Study of lightweight concrete curvature ductility by adding fiber become more complicated.

Lightweight concrete has lower E-modulus compare to normal density concrete and by adding fiber to the concrete it will influence on the E-modulus.

Figure 2-16: Beam section analysis and moment-curvature diagram

(42)

2.15 Concrete fracture mechanical

Ductility can be studied from fracture mechanical parameters such as fracture energy, brittleness number and characteristics length. By study the relationship between the accumulated elastic energy and the fracture energy in the structure can define brittleness number B (bache 1989).

By study Brittleness number formula can understand that ductility of beam is depend on length of beam and maximum stress, when a length or maximum stress increase will get higher brittleness number which means ductility decrease. For increasing ductility in a specific beam length and maximum stress must increase fracture energy and the elastic modulus.

The maximum stress can be replaced by concrete tensile strength to find the brittleness number and by rewriting formula can get the characteristics length too, in that case can brittleness by length.

2 0 3

2

2 0

0

Elastic energy Fracture energy =G

Elastic energy Brittleness number

Fracture energy Where:

is maximum stress L is beam length G is fracture energy E is modulus of e

f

f

f

E L L B L

G E σ α β

α β

σ

σ

=

= =

lasticity

2

2 t f

ch

f t

Lf EG

B l

G E f

= =

(43)

2.16 Fracture energy in concrete

Besides tensile strength and modulus of elasticity, fracture energy should also be in consideration in ductility analysis. The fracture energy can be measured by three point beam bending test.

Fracture energy of concrete can be measured by:

Aggregates quality (size, shape and hardness) and interface between aggregate and mortar are two important parameters which influence the fracture energy of concrete. A research by RILM-TC50 concluded that when the size of aggregate become larger, will fracture energy increase. ref[22] while the interface between aggregates and mortar has direct effect on the tensile properties of concrete.

Strong matrix-aggregates interface increase the area of broken particles, but it is depend on how weak or strong the aggregates are. By having a strong aggregates will the particles just debonded not broken.

In theory can say that because of strong bond and strong aggregates in concrete will crack path wander

0

0

0

( )

is the thikness of beam D- is the ligament length

2

F f

F m

G W

B D a B

a

W W u P

= −

= + ×∆ ∆

Figure 2-17: load displacement curve ref.[22]

(44)

aggregates take place when the interface is strong enough, and by having a weak matrix-aggregates interface will aggregates just be debonded not broken, that will result a lower fracture energy.

Study on strong and weak aggregates shows that fracture energy is first of all depend on how strong is matrix-aggregates interface and than how strong or weak is the aggregates. A various of researches shows that fracture energy of weak aggregates with strong aggregates bond will give lower fracture energy than a strong aggregates with strong aggregate bond ref.[22].

Also for increasing the ductility of lightweight concrete, it should increase the strength of the matrix- aggregates interface which will give increasing in fracture energy in concrete. For reaching the highest value of fracture energy in lightweight concrete, the percentage of broken lightweight aggregates has to be 100%.

Referanser

RELATERTE DOKUMENTER

(a) With overlapping (b) No overlapping Figure 4: The final rendering of the fiber tract cluster from left to right lingual gyrus with both effective connectivity beams.. For a

15 In the temperate language of the UN mission in Afghanistan (UNAMA), the operations of NDS Special Forces, like those of the Khost Protection Force, “appear to be coordinated

For instance, the passive and active network for which the responsibility falls under the network owner and operator, can be denoted as a value network since

Doctor al theses at NTNU , 2021:16Jelena Živković NTNU Norwegian University of Science and Technology Thesis for the Degree of Philosophiae Doctor Faculty of Engineering Department

The strain localization analyses predict a higher ductility for materials with lower yield strength and higher work hardening, as these features lead to a more distributed

In general, the unit cell simulations predict a largely constant tensile ductility with pre-compression, and the strength, work hardening and anisotropy of the materials are found

All the beams had three ø8 mm stirrups in each anchorage zone (behind the supports). Beams without shear reinforcement had just two stirrups in the constant moment region below

In Figure 6b two alternative constructions with wood fiber insulation and 12 mm wood fiber board, one with vapour barrier of PE-foil and the other with a smart vapour barrier